mlsanityData to 14 Oct 2026

Performative Validity of Recourse Explanations

arXiv 2506.15366 ↗First posted 18 Jun 2025stat.ML cs.CY cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

When applicants get rejected by an algorithmic decision system, recourse explanations provide actionable suggestions for how to change their input features to get a positive evaluation. A crucial yet overlooked phenomenon is that recourse explanations are performative: When many applicants act according to their recommendations, their collective behavior may change statistical regularities in the data and, once the model is refitted, also the decision boundary. Consequently, the recourse algorithm may render its own recommendations invalid, such that applicants who make the effort of implementing their recommendations may be rejected again when they reapply. In this work, we formally characterize the conditions under which recourse explanations remain valid under performativity. A key finding is that recourse actions may become invalid if they are influenced by or if they intervene on non-causal variables. Based on our analysis, we caution against the use of standard counterfactual explanations and causal recourse methods, and instead advocate for recourse methods that recommend actions exclusively on causal variables.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 48 of its 63 references that are in the map. First posted on arXiv 18 Jun 2025.

Builds on

48 of 63 references.

References · 48

  1. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,089
  2. 2018Actionable Recourse in Linear Classificationcited by 328
  3. 2020Performative Predictioncited by 327
  4. 2015Strategic Classificationcited by 257
  5. 2020Algorithmic Recourse: from Counterfactual Explanations to Interventionscited by 250
  6. 2019Model-Agnostic Counterfactual Explanations for Consequential Decisionscited by 204
  7. 2020Multi-Objective Counterfactual Explanationscited by 157
  8. 2020Algorithmic recourse under imperfect causal knowledge: a probabilistic approachcited by 140
  9. 2018The Social Cost of Strategic Classificationcited by 105
  10. 2021Towards Robust and Reliable Algorithmic Recoursecited by 102
  11. 2021Outside the Echo Chamber: Optimizing the Performative Riskcited by 96
  12. 2020Stochastic Optimization for Performative Predictioncited by 93
36 more
  1. 2019Strategic Classification is Causal Modeling in Disguisecited by 92
  2. 2018How Do Classifiers Induce Agents To Invest Effort Strategically?cited by 83
  3. 2020On Counterfactual Explanations under Predictive Multiplicitycited by 74
  4. 2021How to Learn when Data Reacts to Your Model: Performative Gradient Descentcited by 71
  5. 2021Strategic Classification in the Darkcited by 66
  6. 2020Causal Strategic Linear Regressioncited by 65
  7. 2020Stochastic optimization with decision-dependent distributionscited by 65
  8. 2021Strategic Classification Made Practicalcited by 59
  9. 2021Who Leads and Who Follows in Strategic Classification?cited by 53
  10. 2023Performative Prediction: Past and Futurecited by 53
  11. 2022Robust Counterfactual Explanations for Tree-Based Ensemblescited by 49
  12. 2021Alternative Microfoundations for Strategic Classificationcited by 47
  13. 2020Maximizing Welfare with Incentive-Aware Evaluation Mechanismscited by 46
  14. 2023Robust Counterfactual Explanations for Neural Networks With Probabilistic Guaranteescited by 45
  15. 2022Performative Powercited by 43
  16. 2022Regret Minimization with Performative Feedbackcited by 40
  17. 2021State Dependent Performative Prediction with Stochastic Approximationcited by 39
  18. 2021A Survey on the Robustness of Feature Importance and Counterfactual Explanationscited by 39
  19. 2021Consistent Counterfactuals for Deep Modelscited by 37
  20. 2020Algorithmic Recourse in the Wild: Understanding the Impact of Data and Model Shiftscited by 36
  21. 2022Making Decisions under Outcome Performativitycited by 33
  22. 2021Strategic Instrumental Variable Regression: Recovering Causal Relationships From Strategic Responsescited by 32
  23. 2019Stabilizing Variable Selection and Regressioncited by 31
  24. 2023Causal Strategic Classification: A Tale of Two Shiftscited by 30
  25. 2021Amortized Generation of Sequential Algorithmic Recourses for Black-box Modelscited by 24
  26. 2019Optimal Decision Making Under Strategic Behaviorcited by 20
  27. 2022Improvement-Focused Causal Recourse (ICR)cited by 20
  28. 2023Provably Robust and Plausible Counterfactual Explanations for Neural Networks via Robust Optimisationcited by 20
  29. 2023Setting the Right Expectations: Algorithmic Recourse Over Timecited by 13
  30. 2023Distributionally Robust Recourse Actioncited by 12
  31. 2024From the Fair Distribution of Predictions to the Fair Distribution of Social Goods: Evaluating the Impact of Fair Machine Learning on Long-Term Unemploymentcited by 12
  32. 2024An engine not a camera: Measuring performative power of online searchcited by 9
  33. 2021A Causal Perspective on Meaningful and Robust Algorithmic Recoursecited by 8
  34. 2023The Risks of Recourse in Binary Classificationcited by 8
  35. 2024Time Can Invalidate Algorithmic Recoursecited by 8
  36. 2025Incentivizing Desirable Effort Profiles in Strategic Classification: The Role of Causality and Uncertaintycited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 7.

8 papers.

  1. 2025Optimal Robust Recourse with Lp-Bounded Model Changecited by 3
  2. 2026ElliCE: Efficient and Provably Robust Algorithmic Recourse via the Rashomon Setscited by 2
  3. 2026Dissecting Performative Prediction: A Comprehensive Surveycited by 2
  4. 2026The Role of Causality in Algorithmic Recoursecited by 1
  5. 2026Performative Learning Theorycited by 0
  6. 2026Off-Policy Evaluation with Strategic Agents via Local Disclosurecited by 0
  7. 2026We Need Explanation Cards to Connect Explanation Algorithms to the Real Worldcited by 0
  8. 2026Algorithmic Recourse Under Competitioncited by 0